UFO format parameters

Asked by NADY on 2011-11-06

Hi Johan

If I have UFO format of any model
and I want to change the mass for any particle How i can do this ?
--> I go to the parameter .py file in the UFO format and change it and this doesn't affect on any other parameters of the particle like width
---> or go to the original file of model then use FeynRules again to produce new UFO format
Cheers,
Nady

Question information

Language:
English Edit question
Status:
Answered
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Last query:
2011-11-09
Last reply:
2011-11-09
Johan Alwall (johan-alwall) said : #1

Hello Nady,

You would normally leave the model files alone and simply change the param_card, which is generated automatically when you output a process. You can also output the param_card by doing
python write_param_card.py
in the model directory.

Note that in order to get physical cross sections for processes with s-channel on-shell particles, you need to ensure that the widths in the param_card are correct, see Question #177457.

All the best,
Johan

--> I go to the parameter .py file in the UFO format and change it and this doesn't affect on any other parameters of the particle like width

Just to add some information on this point: this is perfectly normal, since the information on the parameter are
define in the file restrict_XXXX.dat where XXXX correspond to the model that you have use
the model sm-nobmass will use restrict_nobmass.dat
the model sm-nomasses will use restrict_nomasses.dat
two exception:
the model sm will use restrict_default.dat
the model sm-full will use the value present in the python file.

Cheers,

Olivier

NADY (nady-bakhet) said : #3

Dear Johan,
I'm very Sorry, The problem of change the mass of the particle become very conflict with me
So please give me simple stepes to run it:
I generated this process with MG5 using UFO format for my model
- generate p p > Zprime > e+ e- / /Zprime mass here in UFO is 1500Gev
-output result
-launch result
I get now Mg5/bin/result directroy inside it lhe file with 10000 events
===================================================
Now I want to repeat the previous steps BUT WITH Zprime mass = 1000 Gev to get new lhe file
I must go to parm_card.dat of the MG5/bin/result/card/parm_card.dat
and change Zprime mass to 1000 or what.
after change the mass how i can run the process again
Cheers,
Nady

Hi Nady,

First note than you need to change the mass AND the width! (otherwise the cross-section doesn't make sense).

Concerning the change of the param_card you have mainly two possibilities:

1)
$>./mg5
mg5> launch result
To the question "do you want to edit the param_card say yes, the program will open vi/emacs (depending of your computer and of the default written in the file input/mg5_configuration.txt) and all the rest is the same as the previous run.
If you have already modify it, you are not force to say yes, it will use the current one.

2)
$> cd Mg5/bin/result
$> emacs Cards/param_card.dat
$> ./bin/generate_events

Cheers,

Olivier

NADY (nady-bakhet) said : #5

Hi Olivier,
That is very good, BUT
if the mass was 1500Gev and width is 80
then if i changed the mass to 1000Gev then what the value of new width
Cheers,
Nady

Hi Nady,

We plan to have a module in order to compute the width on the flight.
But at current stage, you need to compute it "by hand". I mean founding yourself all the 1>2 and possibly 1>3 decay mode.
then doing the following:
define 2bodydecay = e+ e- mu+ mu- .......
generate zprime > 2bodydecay 2bodydecay
add process zprime > (all the 3 body decay)
output decay_computation
launch decay_computation
and in the param_card put the mass 1000.
The result will the the total width in GeV.
Then you know what is the width and can put it in the param_card for the cross-section computation.

Cheers,

Olivier

NADY (nady-bakhet) said : #7

Hi Olivier,
Thanks
What about your opinion on this method and please tell me this method is true or false
1-I will open the original file of the model which i defined
2-change the mass of Zprime
3-generate UFO file from Feynrules -----> this will automaticly generate new width
4-use the new UFO in MG5
5-MG5 will gives new results and new cross section due to new UFO which contains new mass and new width
This TRUE or FALSE
Cheers,
Nady

Hi Nady,

This is more a FR question, but to my understanding of the status of FR. FR is not able to compute the width.
In FR, the width is (in most of the models at least) a free parameter.
So this will not work. I know that they plan to compute analytically the width associate to 1 > 2 particles but this is not done for the moment (and they don't plan to include the 1>3 contributions).

Cheers,

Olivier

NADY (nady-bakhet) said : #9

Hi Olivier,
THIS IS YOUR MESSAGE PLEASE ANSWER THE QUESTIONS THAT START WITH ////
===============================
We plan to have a module in order to compute the width on the flight.
But at current stage, you need to compute it "by hand". I mean founding yourself all the 1>2 and possibly 1>3 decay mode.
then doing the following:
define 2bodydecay = e+ e- mu+ mu- .......
generate zprime > 2bodydecay 2bodydecay
add process zprime > (all the 3 body decay)
output decay_computation
launch decay_computation
and in the param_card put the mass 1000. ////At this moment the old value of width in param_card I will leave it without any change or delete it
The result will the total width in GeV. ///Where i can find it "in the web result file" or in param_card.dat
Then you know what is the width and can put it in the param_card for the cross-section computation.//////Do you mean launch decay_computation again after change the width value in parm_card with new value.

Cheers,

Nady

Johan Alwall (johan-alwall) said : #10

Hello Nady,

Some additional comments:
1. 3-body decays should only be included in the decay calculation if there are no 2-body decays. They are typically completely negligible if there are 2-body decays, and since no cuts are applied for a width calculation, you can easily get problems with divergencies in a 3-body decay calculation if you're not careful.
2. Unless you have included the correct width as default value in your FeynRules file, or included the necessary formulas to calculate the width in your FR parameters, I would not recommend to trust the FR width. You should always recalculate the width as described by Olivier
3. For a width calculate, the reported cross section (in pb) corresponds to the width in GeV.
4. Please see Question 177457, where the files needed to get the correct inclusions in the param_card are included.

Best,
Johan

Can you help with this problem?

Provide an answer of your own, or ask NADY for more information if necessary.

To post a message you must log in.